whatmodel.sh / Falcon2 11B (base)
Falcon2 11B (base)
↗ share
+ contribute
Falcon2 11B (base) is an open-weights language model from TII, released May 2024. Its published size is 11.1B parameters and a 8k-token context window. Licence: falcon-llm.
Catalogue identity notes
tiiuae/falcon-11B @a4ab868 (created 2024-05-09, licence falcon-llm). TII Falcon2 11B pretrained model (May 2024; Falcon 2 licence, HF tag 'unknown'). Independent rows: EuroEval English (life_in_the_uk) and Open PT-LLM.
Model information sources
Release date ↗ · Weights licence ↗
HF safetensors metadata 11,102,822,400 parameters.
$ how it compares
The LLMs ranked either side of Falcon2 11B (base), out of 828 with a score, then more from TII.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 592 |  WizardCoder Python 13B V1.0 | 42±2 | local | ||
| 593 |  Hermes 3 Llama 3.1 8B | 42±1 | local | ||
| 594 | Falcon2 11B (base) | 42±3 | local | ||
| 595 |  Granite 34B Code Base (8K) | 42±1 | local | ||
| 596 |  speechless-starcoder2-7b | 42±1 | local | ||
| more from TII | |||||
| 233 | Falcon-H1R 7B | 60±1 | local | ||
| 332 | Falcon-H1 34B Instruct | 54±3 | local | ||
| 494 | Falcon-H1 7B Instruct | 46±3 | local | ||
$ what it's good at
other tests3 not in a task · 95% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Polish EQ-Bench v2 (SpeakLeash) | 42.4% ✎ | ▁··▁▁▂▂▆▅▆█▆ beats 12/70 | 58% |
| Open PT-LLM Leaderboard average (9 Portuguese tasks) | 64.2% ✎ | ▃▁▁▁▁▁▁▁▂█▅▁ beats 108/276 | 22% |
| EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy | 73.7% ✎ | ▁▁▁▁▁▁▁▂▄█▄▃ beats 46/146 | 15% |
$ where to run it
llama.cpp · recommended
version
- Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.macOS / Linux
curl -LsSf https://llama.app/install.sh | shWindowsirm https://llama.app/install.ps1 | iex - Run Falcon2 11B (base) · needs ~6.8 GB · from huggingface.co/bartowskichat in the terminal
llama cli -hf bartowski/falcon-11B-GGUF:Q4_K_M - Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf bartowski/falcon-11B-GGUF:Q4_K_M
LM Studio
version
- Install LM Studio · Open the app once before using the lms command.
- Run Falcon2 11B (base) · needs ~6.8 GB · from huggingface.co/bartowski
- Or download it in the terminal
lms get https://huggingface.co/bartowski/falcon-11B-GGUF@Q4_K_M -y - Then run lms ls to see its name, and lms chat followed by that name
Ollama
- Install Ollama
download it (macOS / Windows) ↗
Linuxcurl -fsSL https://ollama.com/install.sh | sh - Run Falcon2 11B (base) · needs ~6.4 GB · from ollama.comchat in the terminal
ollama run falcon2:11b
Official weights and documentation: huggingface.co/tiiuae ↗
$ how the score was worked out · SpeakLeash Polish EQ-Bench le… 58%, Open Portuguese LLM Leaderboa… 22%, EuroEval leaderboards (result… 15%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 SpeakLeash Polish EQ-Bench leaderboard | 1 | 58% |
| Open Portuguese LLM Leaderboard (archived) | 1 | 22% |
 EuroEval leaderboards (results collection), English generative leaderboard | 1 | 15% |
| prior · pull toward the middle | 5% | |
| weighted average | = 41.8 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json